Using Optical Character Recognition to Identify Legibility of Non-western Languages
Jennifer Owen · DigitalCommons - CalPoly (California State Polytechnic University) · 2011
The purpose of this study is to see if it is viable to use Optical Character Recognition (OCR) to determine legibility of non-western languages.The results will show that OCR can be used to inspect defects in order to waste fewer products.A large amount of waste is an issue in pharmaceutical printing due to the need for products to be one hundred percent accurate.Waste increases with non-western languages when printers do not know that language and cannot determine legibility themselves.The research method used was the scientific method.A paragraph of text taken from pharmaceutical packaging was printed in both English and Kanji.Kanji is a form of Japanese and was used as the test language because of its complex characters.The text was printed using flexography at four different impression levels.Each set of samples was then run through OCR.Results show that the percent wasted was too great using OCR alone.More advanced OCR is recommended and further testing is needed to determine the best combination of OCR and onpress imaging.